Reinforcement Learning for Slate-based Recommender Systems: A Tractable Decomposition and Practical Methodology

Open in new window